發(fā)布時間：2018-08-20 17:51

【摘要】：如今,腸胃疾病已經(jīng)成為人類健康的一大威脅。機械式內(nèi)窺鏡作為傳統(tǒng)腸胃疾病的檢測手段,不僅操作不方便,也給患者帶來身體上的痛苦。隨著半導體技術(shù)、無線通信技術(shù)、集成電路技術(shù)等技術(shù)的發(fā)展,膠囊內(nèi)窺鏡問世了,并很快在腸胃疾病的檢測方面取得得天獨厚的優(yōu)勢。然而膠囊內(nèi)窺鏡一次檢測產(chǎn)生的消化道圖像數(shù)量達幾萬張之巨,由醫(yī)護人員來逐一查看,給醫(yī)務(wù)人員帶來沉重負擔的同時,還會增加誤診率。針對這些問題,本文在介紹了膠囊內(nèi)窺鏡和基于膠囊內(nèi)窺鏡圖像的出血病灶檢測技術(shù)的國內(nèi)外研究現(xiàn)狀的基礎(chǔ)上,對基于膠囊內(nèi)窺鏡圖像的出血病灶檢測技術(shù)進行了深入的研究,主要的工作包括:圖像的預處理、基于顏色特征的感興趣區(qū)域提取、基于顏色相似性和連通域面積的分類識別,實現(xiàn)了基于膠囊內(nèi)窺鏡圖像的出血病灶檢測算法。傳統(tǒng)的基于膠囊內(nèi)窺鏡圖像的出血檢測算法主要有兩種,一種是將圖像分割為固定大小的區(qū)域,這種機械的劃分會破壞圖像本身含有的邊界信息,導致準確度不高;另一種是在整幅圖像上進行模板運算,這種算法可以最大限度反映原始圖像的信息,然而由于數(shù)據(jù)量大,導致檢測算法速度過慢。本文兼顧算法的速度和圖像的原始邊界信息,首先在RGB顏色空間利用顏色邊界盒進行感興趣區(qū)域提取,減少圖像的冗余信息,然后利用感興趣區(qū)域的顏色相似性系數(shù)和連通域面積組成的分類器,對感興趣區(qū)域進行分類,在保證檢測準確率的前提下提高了檢測速度。最后,本文通過實驗驗證算法,結(jié)果表明:算法的靈敏度達到了91%,特異性達到88%,基本實現(xiàn)了膠囊內(nèi)窺圖像出血病灶的自動檢測,可應(yīng)用于實踐治療中。
[Abstract]:Nowadays, gastrointestinal diseases have become a major threat to human health. Mechanical endoscopy, as a traditional method of detecting gastrointestinal diseases, is not only inconvenient to operate, but also brings physical pain to patients. With the development of semiconductor technology, wireless communication technology, integrated circuit technology and so on, the capsule endoscope came out, and soon obtained the unique advantage in the detection of gastrointestinal diseases. However, the number of digestive tract images produced by a capsule endoscope is tens of thousands of Zhang Zhi, which is examined by medical staff one by one, which brings a heavy burden to medical staff and increases the misdiagnosis rate at the same time. Aiming at these problems, this paper introduces the research status of capsule endoscopy and hemorrhage focus detection technology based on capsule endoscopy image at home and abroad. Based on the capsule endoscope image, the detection technology of hemorrhage focus is studied in depth. The main work includes: image preprocessing, extraction of region of interest based on color feature, classification and recognition based on color similarity and area of connectivity. An algorithm for detecting hemorrhage focus based on capsule endoscopy image is implemented. There are two kinds of traditional hemorrhage detection algorithms based on capsule endoscope image. One is to divide the image into a fixed size area. This kind of mechanical partition will destroy the boundary information contained in the image itself and lead to low accuracy. The other is template operation on the whole image, which can reflect the information of the original image to the maximum extent. However, because of the large amount of data, the speed of the detection algorithm is too slow. In this paper, the speed of the algorithm and the original edge information of the image are taken into account. Firstly, the region of interest is extracted by using the color boundary box in the RGB color space to reduce the redundant information of the image. Then the region of interest is classified by using a classifier composed of the color similarity coefficient of the region of interest and the area of the connected domain. The detection speed is improved on the premise of ensuring the detection accuracy. Finally, the algorithm is verified by experiments. The results show that the sensitivity of the algorithm reaches 91%, the specificity reaches 88%, and the automatic detection of hemorrhage focus in capsule endoscope image is basically realized, which can be applied to practical treatment.
【學位授予單位】：華中科技大學
【學位級別】：碩士
【學位授予年份】：2016
【分類號】：R57;TP391.41

【參考文獻】

相關(guān)期刊論文 前10條

1 滕秀花;胡文瑜;陳敏;;一種基于SLIC的超像素快速色彩傳遞算法[J];哈爾濱師范大學自然科學學報;2014年03期

2 石煜;顏國正;朱柄全;;視頻膠囊內(nèi)窺鏡無線能量接收系統(tǒng)的設(shè)計[J];儀器儀表學報;2014年03期

3 林睿;王邦茂;樸美玉;;OMOM膠囊內(nèi)鏡在小腸疾病診斷中的應(yīng)用[J];江蘇醫(yī)藥;2013年19期

4 孫賢久;李學謙;秦先鋒;蘇劍東;劉美紅;張莉;江堤;劉玉杰;朱惠明;;OMOM膠囊內(nèi)鏡對小腸疾病的診斷[J];吉林醫(yī)學;2013年27期

5 賈智偉;顏國正;石煜;樊紹勝;;生理參數(shù)遙測系統(tǒng)無線供能模塊的設(shè)計[J];納米技術(shù)與精密工程;2014年02期

6 王曉麗;趙潔;馬阿火;;小腸疾病123例OMOM膠囊內(nèi)鏡診斷結(jié)果分析[J];中國鄉(xiāng)村醫(yī)藥;2013年18期

7 曲寶戈;王慧;潘錦敦;喬瑞玲;任光鶯;;OMOM膠囊內(nèi)鏡在老年人胃和小腸檢查中的應(yīng)用[J];胃腸病學和肝病學雜志;2013年06期

8 付延安;孟慶虎;張偉;;基于超像素分割的無線內(nèi)窺鏡出血圖像檢測[J];吉林大學學報(工學版);2013年02期

9 潘國兵;顏國正;宋昕帥;邱祥玲;;膠囊內(nèi)窺圖像出血檢測中顏色向量相似系數(shù)分類器的設(shè)計[J];上海交通大學學報;2009年11期

10 謝翔;李國林;張春;王志華;;一種雙向、數(shù)字式微型無線內(nèi)窺鏡系統(tǒng)設(shè)計[J];固體電子學研究與進展;2007年01期

相關(guān)碩士學位論文 前1條

1 王微微;體外磁驅(qū)動診療膠囊系統(tǒng)研究[D];上海交通大學;2013年

，

本文編號：2194502